Lee et al v. Virginia State Board of Elections, et al
Filing
223
MEMORANDUM OPINION. Signed by District Judge Henry E. Hudson on 05/19/2016. (tjoh, )
r:j
I
;1
IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF VIRGINIA
Richmond Division
i'.
MAY 1 9 20IG
J
PI
i^-RK
iis
BARBARA H. LEE, etal.
Plaintiffs,
Civil Action No. 3:15CV357-HEH
V.
VIRGINIA STATE BOARD OF
ELECTIONS, etal
Defendants.
MEMORANDUM OPINION
This is an action challenging, on a number of fronts, the constitutionality of
Virginia Code § 24.2-643(B), commonly referred to as the Virginia voter ID lav^ or
Senate Bill 1256 ("SB 1256"). In effect, this statutory provision requires voters in the
Commonwealth of Virginia to present a form of statutorily-approved identification in
order to vote. The approved forms of identification include a valid Virginia driver's
license, U.S. passport, or other photo identification issued by the Commonwealth of
Virginia, one of its political subdivisions, or the United States; a valid student
identification card containing a photograph of the voter and issued by any institution of
higher education located in the Commonwealth; or any valid employee identification card
containing a photograph of the voter and issued by an employer of the voter in the
ordinary course of the employer's business. Va. Code § 24.2-643(B);' see also Pis.' Trial
' Other portions ofthe Amended Complaint seeking redress for long lines at polling precincts
and automatic re-enfranchisement of persons convicted of non-violent felonies were either
dismissed or resolved by the parties.
, r
Ex. 151 - Voter Identification Chart.
Voters who are unable to produce valid identification are permitted to cast a
provisional ballot which must be cured by the Friday succeeding election day. When
voters are provided with a provisional ballot for lack of properidentification, it is noted
by the election official in the provisional ballot log. The notation specifically includes
that no identification is the reason the person is casting a provisional ballot. The voter is
then advised of the procedure to cure, enabling her ballot to be counted. To cure a
provisional ballot, voters must present valid identification to the local registrar either in
person or by fax or email. (Trial Tr. 945:14-46:15, Feb. 25, 2016 (Test, of Myron
McClees).)
Under the regulations implementing SB 1256, voters without valid identification
can obtain free photographic voter identification at a local registrar's office. The
application process requires the voter to identify herself by date of birth and social
security number. After confirming that the applicant is a registered voter, her picture is
taken and her signature is recorded on a digital pad. A photograph-bearing identification
card is then sent to the voter's address of record free of charge. This form of voter
identification can only be issued at registrar's offices because it requires access to a
secure computer system containing the voter's personal identification. Consequently,
such identification cannot be issued at polling stations. {Id. at 1449:9-50:7,
1465:16-66:11, Feb. 26, 2016 (Test. ofEdgardo Cortes).)
Plaintiffs urge the Court to issue a permanent injunction enjoining the
Commonwealth of Virginia and its agents from enforcing the voter ID law. In addition,
2
they ask the Court to find that the photo ID requirement for voting adversely impacts
minority voters in violation of Section 2 of the Voting Rights Act, as well as the First,
Fourteenth and Fifteenth Amendments of the Constitution of the United States. Plaintiffs
also ask this Court to find that the voter ID law intentionally discriminates against young
voters in contravention of the Twenty-Sixth Amendment.
Following resolution of pretrial motions addressing Plaintiffs' standing under
Federal Rule of Civil Procedure 12(b)(1) and the adequacy of the underlying allegations
under Rule 12(b)(6), this Court conducted a seven day trial without a jury. At the close
of the evidence, in lieu of oral argument, the Court afforded each party an opportunity to
file post-trial memoranda supporting their respective positions with specific references to
pertinent portions of the voluminous documents placed into evidence in this case.^ This
opinion followed.
The core contention in this case is that the voter identification law was enacted by
the Virginia General Assembly with the intention of gaining partisan advantage by
placing an undue burden on certain classes of opposition voters. Count I alleges a
violation of Section 2 of the Voting Rights Act. In support, Plaintiffs contend that the
voter ID law has an adverse disparate impact on Afncan American and Latino voters.
Plaintiffs maintain that the law imposes a discriminatory burden on a protected class,
fostered in part by social and historical conditions in the Commonwealth of Virginia. In
Count II, Plaintiffs maintain that the Virginia voter ID law violates the First Amendment
^The parties introduced approximately 8,000 pages ofdocuments inthis case, plus an assortment
of video tapes.
3
and the Equal Protection Clause of the Fourteenth Amendment in that it imposes an
undue burden on the right to vote and results in disparate treatment of protected classes
without a rational basis. Count III, styled "Partisan Fencing," alleges that the Virginia
voter ID law subjects a group of voters to disfavored treatment by reason of their political
views. Lastly, Counts IV and V allege intentional discrimination by race and age,
respectively. Plaintiffs contend that the Virginia General Assembly enacted the Virginia
voter ID law with the specific intent to suppress African American, Latino, and young
voters.
In part, Plaintiffs' evidence consisted of testimony from a dozen Virginia voters
who alleged that they were burdened by the Virginia voter ID law in casting their ballots
during the 2014 and 2015 election cycles. These individuals cited a variety of
impediments that allegedly made the voting process unduly cumbersome. But in most
cases, complying with the law proved to be a surmountable hurdle. Plaintiffs offered a
variety of expert witnesses describing the history of racial discrimination in Virginia
politics and a demographic breakdown and analysis of segments of the Virginia
population who may not possess valid identification. Virginia election officials and
members of the General Assembly provided some legislative history on the enactment on
the Virginia voter ID law and its implementation by the Virginia State Board of Elections
("SBOE").
The centerpiece of Plaintiffs' evidence was the expert testimony of several
professors with extensive experience testifying in election law related cases. One expert
concluded that in person voter identification fraud was rare. Another, after providing an
4
overview of election laws adopted by approximately thirty other states, concluded that
there was no rational basis for the adoption of the Virginia voter ID law, and given the
history of discrimination in Virginia, must have been adopted for the purpose of
suppressing minority votes.
The Defendants countered with a number of expert witnesses who pointed out that
the statistical analysis employed by Plaintiffs arguably omitted a large segment of
Virginia voters who likely would have valid identification. Defendants' experts also
testified that based on their investigation and analysis, the implementation of the
Virginia voter ID law resulted in very few individuals being unable to casta vote during
the 2014 election cycle. They described the burden imposed by the Virginiavoter ID law
as having a fairly even effect on individuals of all ages, races, and nationalities.
Furthermore, under the statutory scheme adopted under SB 1256, no voterwas actually
disenfranchised; each had a means of casting a ballot if he or she chose to exercise
alternative voting options.
Lastly, the Defendants' experts pointed out that while the number of actual
convictions for voter fraud may be minimal, that statistic may not accurately reflect the
number of such cases reported to law enforcement authorities. Irrespective of statistics,
one defense expert testified that in her opinion, several legitimate reasons existed
warranting passage of the voter ID law, including that a large segment of the Virginia
population had a perception that in person voter fraud could potentially occur and
supported the legislation challenged in this case.
To provide some insight into the deliberative process underlying the enactment
5
and the implantation of SB 1256, the Defendants, a former Secretary of the SBOE, and a
number of SBOE and Virginia Department of Elections ("VDOE") employees, along
with other local election officials, outlinedthe extensive public information campaign
launched in 2014 to educate voters on the necessity for proper identificationwhen voting.
Edgardo Cortes ("Cortes"), the current Commissioner of the VDOE, who opposed SB
1256, testified that he and his staff attempted to implement the law in the least
burdensome way possible. (Trial Tr. 1500:11-18, Feb. 26, 2016.) While this outreach
was not flawless, it included a large swath of voters.
The evidence in this case clearly demonstrated, as both parties will concede, that
Virginia has an unfortunate history of racial discrimination and statutory artifice to hinder
black voting. The evidence is equally clear that prior to the adoption of the Voting
Rights Act in 1965 ("the Act"), legislation was enacted by the Virginia General
Assembly that materially affected the rights of African Americans to vote. The Voting
Rights Act was intended as a safeguard against policies and practices undermining an
equal opportunity by black and white voters alike to elect their preferred representatives.
Thornburg v. Gingles, 478 U.S. 30,47 (1986). While the Act undoubtedly ushered in
significant reform measures, underlying issues continued to spark partisan debate.
The evidence also revealed that the Virginia voter ID law has created a layer of
inconvenience for some voters. But the question squarely presented in this case is
whether Virginia Code § 24.2-643(B) is unconstitutional either in its adoption,
implementation, or enforcement. Does it, by design or otherwise, adversely affect the
opportunity of minorities to vote or is the burden evenly spread? Is a legislative body's
6
authority to adopt protective legislation dependent on objective criteria or their delegated
judgment?
I.
A Second Look at Standing
Before turning to the merits of Plaintiffs' claims, the Defendants urge the Court to
revisit its earlier finding that the Democratic Party of Virginia ("DPVA") has Article III
standing. In a Memorandum Opinion issued December 18, 2015, this Court concluded,
based upon a facial review of the Amended Complaint, that "[i]n the immediate case, the
DPVAclaims direct injury to its raison d'etre—electing candidates who supportthe
Democratic platform, as opposed to individualized interests of its members." (Mem. Op.
8, ECFNo. 110).
The testimony at trial appears to support this conclusion. While it has no formal
membership roster, the DPVA is an umbrella organization encompassing committees of
supporters in every city and county in Virginia. Rebecca Slutzky ("Slutzky"), Executive
Director of the DPVA, testified that under the party plan, it includes anyone who leans
Democratic, votes Democratic, or supports the Party. As the United States Supreme
Court explained in Warth v. Seldin, 422 U.S. 490 (1975), associations can allege standing
based upon two distinct theories. First, the association "may have standing in its own
right to seek judicial relief from injury to itself and to vindicate whatever rights and
immunities the association itself may enjoy." Warth, 422 U.S. at 511. Second, the
association may have standing as the representative of its members who have been
harmed. Id.; see also Hunt v. Wash. State Apple Advert. Comm % 432 U.S. 333, 342^3
(1977).
7
The DPVA serves as an umbrella organization overseeing local committees
composed of Democratic supporters, coordinating statewide campaign strategies, and
promoting voter turnout. The party expended time and resources to educate voters and
party members on the requirements of SB 1256. The party obtained lists of individuals
who may not possess proper identification for information targeting. It also hired a voter
protection director whose responsibilities included the identification and education of
voters potentially burdened by identification requirements. Slutzky also testified that the
voter education program necessitated by SB 1256 detracted time and resources that
would have otherwise been expended increasing voter turnout.
Both the chair and vice chair of the Henrico County Democratic Committee
described similar experiences. Both were active in voter identification education. Cheryl
Zando ("Zando"), Chair of the Henrico County Democratic Committee, also chaired a
task force which organized phone banks promoting free identification available at the
registrar's office. Cathy Woodson ("Woodson"), Vice Chair of the Henrico County
Democratic Committee, organized outreach projects at community events to familiarize
voters with identification requirements and access to free forms of valid identification.
Both Zando and Woodson testified that but for the need to educate voters on the
requirements of SB 1256, they would have engaged in other campaign-related activity.
Near identical experiences were recounted by Susan B. Kellom, Chair of the Alexandria
City Democratic Committee, and Jeff Allen ("Allen"), a Democratic Party field
organizer.
Collectively viewed, the DPVA has shown sufficient injury primarily inthe form
8
of diversion of time, talent, and resources to educate their voters and implement the
requirements of the Virginia voter identification law. See Crawford v. Marion Cnty.
Elec. Bd., 472 F.3d 949, 951 (7th Cir. 2007) (citing Friends ofthe Earth, Inc. v. Laidlaw
Envtl. Sevrs. Inc., 528 U.S. 167, 180-84 (2000)), affirmed 552> U.S. 181 (2008)).
In the Court's opinion. Plaintiffs have satisfiedtheir burden of demonstrating a
realistic danger of sustaining direct injury as a result of SB 1256, if in fact it suppresses
minority voters likely to support Democratic candidates. Fla. State Conf ofNAACP v.
Browning, 522 F.3d 1153, 1161 (11th Cir. 2008). Both individual Plaintiffs in this case
are registered voters in Virginia who affiliate themselves with the Democratic Party.
They express an intention to vote for Democratic candidates in the future and have been
involved in voter registration, education, and voter turnout projects. Both Barbara H. Lee
("Lee") and Gonzalo Aida Brescia ("Aida") are members of their local Democratic
committee and intendto participate in get-out-the-vote activities during the next election
cycle. Aida also testified that as a result of the enactment of SB 1256, he had the
additional burden of preparing educational materials on valid forms of voter
identification, including emails, graphics, and Facebook postings. These tasks consumed
time that he would have otherwise devoted to issue and candidate advocacy.
11.
Overview of Legal Standards by Which Evidence is Measured
Section 2 of the Voting Rights Act, codified at 52 U.S.C. § 10301(a), prohibits any
"standard, practice, or procedure ... which results in a denial or abridgement of the right
of any citizen of the United States to vote on account of race or color." The statute
further explains that "'[a] violation of subsection (a) is established if, based onthe totality
of the circumstances, it is shown that the political processes leading to nomination or
election in the State or political subdivision are not equally open to participation by'
citizens ofprotected races 'in that [they] have less opportunity than other members of the
electorate to participate in the political process and to elect representatives of their
choice.'" League of Women Voters ofN.C. v. North Carolina, 769 F.3d 224, 238 (4th
Cir. 2014) (second alteration in original) (quoting 52 U.S.C. § 10301(b)).
The central inquiry under Section 2 "is whether as a result of the challenged
practice or structure plaintiffs do not have an equal opportunity to participate in the
political processes and to elect candidates of their choice." Gingles, 478 U.S. at 44
(internal quotation marks and citation omitted). Proof of intentional discrimination is
unnecessary to prevail on a Section 2 claim. Proof of discriminatory results is sufficient.
Chisom V. Roemer, 501 U.S. 380, 404 (1991). "The essence of a [Section] 2 claim is that
a certain electoral law, practice, or structure interacts with social and historical conditions
to cause an inequality in the opportunities enjoyed by black and white voters to elect their
preferred representatives." Gingles, 478 U.S. at 47.
Based on a thorough analysis of Section 2 vote-denial jurisprudence, the Court of
Appeals for the Fourth Circuit in League of Women Voters ofNorth Carolina isolated the
two critical elements of proving such a claim:
First, the challenged standard, practice, or procedure must impose a
discriminatory burden on members of a protected class, meaning that
members of the protected class have less opportunity than other members
of the electorate to participate in the political process and to elect
representatives of their choice. Second, that burden must in part be caused
by or linked to social and historical conditions that have or currently
produce discrimination against members of the protected class.
10
769 F.3d at 240 (internal quotation marks and citations omitted).
The Supreme Court has continually counseled that vote-denial cases brought under
Section 2 should not be viewed in isolation, but should be evaluated in light ofthe totality
of circumstances. The court in Gingles suggested a number of potentially relevant
factors. These include: (1) any history of voting-related discrimination inthe pertinent
state; (2) the extent to which voting isracially polarized; (3) the history of use of voting
practices or procedures that tend to enhance the opportunity for discrimination against
minority groups; (4) the exclusion of members of the minority group from candidate
slating processes; (5) the extent to which minority group members bear the effects of past
discrimination in areas such as education, employment, and health, which hinder their
ability to participate effectively in the political process; (6) the use of even subtle racial
appeals in political campaigns; (7) the extent to which the members of the minority group
have been elected to public office in the jurisdiction; (8) evidence that elected officials
are unresponsive to the particularized needs of members of the minority group; and (9)
the extent to which the policy underlying the state's use of the practice or structure at
issue is tenuous. Gingles, 478 U.S. at 44-45.
In applying the analytical framework articulated in Gingles, "'there is no
requirement that any particular number of factors be proved, or [even] that a majority of
them point one way or the other.' ... Instead, courts must undertake 'a searching
practical evaluation of the past and present reality,' [with] a 'functional' view of the
political process." League of Women VotersofN.C., 769 F.3d at 240-41 (alterations in
11
original) (quoting Gingles, 478 U.S. at 45).
Turning to the First Amendment and Equal Protection claims raised in CountII of
the Amended Complaint, this Court's review is guided by the balancing framework
articulated in Anderson v. Celebrezze, 460 U.S. 780 (1983), and amplified by Burdickv.
Takushi, 504 U.S. 428 (1992). The Court succinctly framed the Anderson-Burdick
controlling standard in Crawford v. Marion County Election Board: "a court evaluating a
constitutional challenge to an election regulation [must] weigh the asserted injury to the
right to vote against the precise interests put forward bythe State asjustifications for the
burden imposed by its rule." 553 U.S. 181, 190 (2008) (quoting Burdick, 504 U.S. at
434) (internal quotation marks omitted).
Finally, the teachings of the Supreme Court in Village ofArlington Heights v.
Metropolitan Housing Development Corp., 429 U.S. 252 (1977), are instructive in
analyzing the intentional discrimination claims in Counts IV and V. The court in
Arlington Heights restated the well-established tenet that "[p]roof of racially
discriminatory intent or purpose is required to show a violation of the Equal Protection
Clause." Id. at 265. Arlington Heights identified a number of factors to be employed by
reviewing courts in evaluating facially neutral laws allegedly passed with a
discriminatory purpose. This evaluation requires courts to perform "a sensitive inquiry
into such circumstantial and direct evidence of intent as may be available." Id. at 266.
The court fiirther stressed that the impact of the official action may provide an important
starting point under discriminatory purpose analysis. Id.
In assessing whether racial discrimination has been demonstrated to be a
12
substantial or motivating factor behind the enactment oflegislation, Arlington Heights
also delineated a number of non-exhausting factors to guide the court: (1) the historical
background of the decision-making process, particularly if it indicates a series of official
actions taken for invidious purposes; (2) the specific sequence ofevents leading up to the
challenged legislative action; (3) departures from normal procedural sequence; (4)
substantive departures, particularly if the factors usually considered important by the
decision-maker strongly favor a decision contrary to the one reached; (5) the legislative
or administrative history especially where they are contemporary statements by members
of the decision-making body, minutes of its meetings or reports. Id. at 267-68.
"Once racial discrimination is shown to have been a substantial or motivating
factor behind the enactment of the law, the burden shifts to the law's defenders to
demonstrate that the law would have been enacted without this factor." Hunter v.
Underwood, 471 U.S. 222, 228 (1985) (internal quotation marks and citation omitted).
III.
Legislative History of Virginia Voter Identification Bills
To provide historical context for the present litigation, some explanation of the
evolution of SB 1256 may serve as an enlightening preface. The stage is set with the
adoption of the Help America Vote Act ("HAVA") of 2002 by the United States
Congress. With the objective of protecting the integrity of the electoral process, HAVA
imposed a number of requirements on the individual states. Congress required every
state to create and maintain a computerized statewide list of all registered voters. 52
U.S.C. § 21083(a)(1)(A). In addition, HAVA required states to verify voter information
contained on a voter registration application by using the applicant's driver's license
13
number or the last four digits of the applicant's social security number. Id. §
21083(a)(5)(A)(i). Most pertinent to the case at hand,
HAVA also impose[d] new identification requirements for individuals
registering to vote for the first time who submit their applications by mail.
If the voter is casting his ballot in person, he must present local election
officials with written identification, which may be either "a current and
valid photo identification" or another form of documentation such as a bank
statement or pay check. If the voter is voting by mail, he must include a
copy of the identification with his ballot. A voter must also include a copy
of the documentation with his application or provide his driver's license
number or Social Security number for verification. Finally, in a provision
entitled "Fail-safe voting," HAVA authorizes the casting of provisional
ballots by challenged voters.
Crawford, 553 U.S. at 193 (citations omitted).
The photograph identification requirements established by HAVA applied only to
federal elections. However, as the Court pointed out in Crawford, the safeguard
measures adopted in HAVA did "indicate that Congress believes that photo identification
is one effective method of establishing a voter's qualification to vote and that the
integrity of elections is enhanced through improved technology. That conclusion is also
supported by ... the Commission on Federal ElectionReform, chaired by former
President Jimmy Carter and former Secretary of State James A. Baker III." Id.
The requirement that voters present a non-photo form of identification at the polls
has been in effect since 1996 in the Commonwealth of Virginia. A registered voter
without any form of identification could cast his or her ballot by simply executing an
affirmation of identity. (Trial Tr. 956:20-24, Feb. 25, 2016 (Test, of Myron McClees).)
In 2012, it became apparentto the SBOE that the mere signing of an affirmation of
identity for first time voters in federal elections who registered bymail was inadequate to
14
comply with HAVA standards. {Id. at 1611:15-12:4, Mar. 1, 2016 (Test, of Donald
Palmer).) The differing identification procedures for state and federal elections created
considerable confusion among poll workers. (Id. at 1611:18-12:11.)
In 2012, the Virginia General Assembly rescinded the self-affirmation procedure
and substituted a limited field of non-photograph bearing identification. To promote
uniformity, voters in Virginia were required to produce one of the specified forms of
identification in all elections, both state and federal. Va. Code § 24.2-643 (version
effective until July 1, 2014). Voters could register by mail without submitting any form
of identification and receive a non-photo registration card. While largely supported by
Republicans, this legislation was adopted with bipartisan support. It was subsequently
reviewed and precleared by the United States Department of Justice as required by the
Voting Rights Act.
In 2013, just one year aftera voter identification bill had been adopted, the
General Assembly passed SB 1256. While this legislation expanded the list of
permissible forms of identification, it required that the identification include a photograph
ofthe voter. The bill, introduced by Senator Mark Obenshain ("Senator Obenshain"),
sparked spirited partisan debate on the floor of the Virginia General Assembly. SB 1256
was ultimately adopted with unanimous Republican support, coupled with one Democrat
and one Independent supporter. (Trial Tr. 1615:8-14, Mar. 1,2016 (Test, of Donald
Palmer).) The law as enacted also provided for the issuance ofa free photograph-bearing
voter identification card by local registrars' offices. If the applicant is a registered voter,
no further identification is required to obtain a free photo ID. Va. Code § 24.2-643
15
(version effective from July 1, 2014). Approximately 4,500 free photo IDs have been
issued.
The language of SB 1256 also required that the photo ID be valid. Donald L.
Palmer ("Palmer"), Secretary of the SBOE in 2013, testified that the definition of the
term "valid" kindled considerable debate between the SBOE and Senator Obenshain,
patron of SB 1256. In reviewing SBOE's regulations, Palmer discovered that in either
2000 or 2001, the SBOE had issued guidance to registrars that any form of identification
expired in excess of thirty days shouldbe considered invalid. In the ensuing discussions
regarding SB 1256's implementation, the SBOE was not in favor of adopting any specific
expiration period. Senator Obenshain sent a letter to the SBOE challenging its authority
to permit any form of expired identification to be honored by election officials as valid.
The SBOE, over the Senator's objection, voted to define a valid identification as one not
expired over one year. (Trial Tr. 1621:4-17, Mar. 1,2016(Test, of Donald Palmer).)
After inviting public comment, the regulation was adopted. Palmer also testified
that in his opinion, SB 1256 deterred voter fraud and served as a valuable safeguard. (Id.
at 1634:5-7.) In fact, he recalled that a computerized interstate cross-check of persons
voting in Virginia against votes cast in other states revealed several cases of possible
multiple voting. These cases were referred to the Virginia State Police but did not result
in prosecution. (M at 1682:3-23.)
According to Palmer, many of the provisions of SB 1256 were modeled after voter
ID laws adopted in other states such as Georgia and South Carolina, which had been
precleared by theDepartment of Justice pursuant to the Voting Rights Act. (Id. at
16
1650:16-23, 1680:10-14.)
J. Justin Riemer ("Riemer"), Deputy Secretary of the SBOE from October 2011
through January 2014, testified to several initiatives the SBOE undertook to improve the
electoral process in Virginia. For example, the SBOE promoted legislation allowing for
the Department of Motor Vehicles ("DMV") to transmit completed voter registration
forms electronically to the appropriate registrar's office so that those voters may be
registered and added to the voter rolls. {Id. at 1554:3-55:11.) Additionally, the SBOE
attempted to improve the process for absentee voting by allowing voters to apply for an
absentee ballot online. {Id. at 1555:17-56:3.) Although this initiative ultimately
launched after his tenure with the SBOE, he helped to lay the policy groundwork for its
implementation. {Id. at 1556:4-6.)
In the debate preceding the adoption of SB 1256, Riemer recalled commentary in
the General Assembly concerning the existence of voter fraud. He specifically
remembered a comment by Senator Thomas A. Garrett ("Senator Garrett"), in his former
capacity as a Commonwealth's Attorney, that Garrett had prosecuted such a case. He
also remembered an article in the Richmond Times-Dispatch indicating that voterrelated
fraud may be a "bigger problem in Virginia than [the Times-Dispatch] had realized and
... had acknowledged." {Id. at 1563:11-64:2.) While Riemer recalled reports of voter
registration fraud, he admitted no knowledge of any prosecution for in person voter fraud.
He noted that the SBOE conducted no formal study of voter fraud before SB 1256 was
adopted. {Id. at 1573:6-9.)
Riemer did testify that an analysis was conducted to determine how many voters
17
were in the DMV's system as either having operator's licenses or other forms of DMV
identification. The results indicated that 93.22 percent of active voters in Virginia had
some form of DMV-issued identification. (Pis.' Trial Exs. 168, 185.) Riemer conceded
though that this statistic did not reflect the number of individuals residing in rural areas
without access to a motor vehicle or were too disabled to get to polling locations.
The 2012 legislation, in its original form, included a provision requiring voters to
presentphoto identification. Delegate Jennifer McClellan ("DelegateMcClellan"), a
Democrat representing the Richmond area, testified that she was so concerned about its
effect on her minority constituents that she approached Governor Robert McDonnell, a
Republican, for assistance. Delegate McClellan described her district as an economically
diverse majority-minority district with a total black population approaching seventy
percent. She also believed that a number of her constituents bom as late as 1940s may
not have birth certificates enabling them to acquire the necessary identification.
Furthermore, in her opinion, the photo identification issued by the DMV was the
equivalent of a poll tax because of the $10 cost. She testified that she found the
Republican rationale for the photograph bearing identification to be unpersuasive. She
was unaware of any reported incidents of voter fraud that would be deterred by such
legislation. (Trial Tr. 376:14-77:8, Feb. 23, 2016.)
Delegate McClellan convinced Governor McDonnell that the photograph
requirement would place an undue burden onher less-affluent constituents. At Delegate
McClellan's urging, and after conferring with other groups representing minority
interests. Governor McDonnell amended the 2012 legislation by adding non-photo ID
18
options to the list of acceptable forms of identification. Governor McDonnell also
pressed the General Assembly to include a budget item underwriting the cost of
educating voters on the new identification requirements. Despite these modifications to
the legislation, Delegate McClellan still opposed the 2012 voter identification bill. She
testified that she had many constituents who were unemployed, had no driver's license,
or any form of student identification. (Id. at 377:9-18.) Moreover, she stridently
opposed the 2013 bill which revived the photo identification requirement. She added that
no Afncan-American member of the GeneralAssembly supportedthe 2013 bill which, in
her view, burdened her constituency. Voters in her district were opposed to the 2013
voter identification law because there was no compelling reason to amend the 2012 law
by adding a photo identification requirement. {Id. at 380:18-82:7.) In their view,
nothing occurred between 2012 and 2013 to justify such action.
Prior to being elected to the Virginia Senate in 2015, Scott A. Surovell ("Senator
Surovell") represented the Mount Vemon area of Fairfax County in the House of
Delegates. He described his House district as predominately upper class with the
exception of Gum Springs, an historic area with a lower income mix of African American
and Latino population. The Senator described himself as a life-long political activist
aggressively involved in voter recruitment and working the polls. Senator Surovell
testified that he dedicated a considerable amount of time as a House member interacting
personally with Gum Springs constituents. In his campaign for the state Senate in 2015,
Senator Surovell testified that he knocked on approximately 25,000 doors in the area he
represents.
19
Although Senator Surovell had only anecdotal evidence, and minimal hard
numbers, he suspected that many Gum Springs residents had neither the resources nor the
transportation to obtain any form of valid photo identification. Many residents of that
area had no Virginia driver's license and relied on public transportation. The Fairfax
County Registrar's Office, according to Senator Surovell, is located in the government
center, which is approximately a two-hour bus ride from his district, and a forty-five
minute commute by car. He believed this distance made a free form of voter
identification beyond the reach of some lower income voters.
Despite an unscientific poll of residents of his House district narrowly favoring a
photo identification requirement to vote,^ Senator Surovell led the opposition to such
legislation onthe House floor. Hetoo was unconvinced that there were any reported
incidents of voter impersonation in Virginia warranting such legislative action. He
remembered asking his Republican colleagues to offer examples of voter fraud. He
recalls none. In his view, the 2013 legislation was a "solution in search of a
problem." (Jd. at 312:23-24.) Although Senator Surovell argued forcefiilly onthe floor
ofthe Virginia General Assembly that such legislation limited the constitutional right of
his constituents to vote, he admitted that he was unaware of any incidents where someone
was actually denied the right to vote as a result of the photo identification law. While
Senator Surovell suspected partisan motives for the adoption ofthe 2013 legislation, he
conceded thatpopular support for photo identification was probably a factor in its
In Senator Surovell's informal survey of his constituents, he received between 400 and 600
responses which he described as"either an even split, ora slight majority in favor of ID." (Trial
Tr. 359:6-60:4, Feb. 23, 2016.)
20
adoption.
Algie Howell, Jr., a former member of the Virginia House of Delegates, currently
serving on the Virginia Parole Board, opposed voter identification laws. His opposition
was based in part on his personal experience attending racially-segregated schools in
Virginia, and what he described in his testimony as Virginia's fifty-year history of
discrimination. He recalled that many members of his African-American family had no
education. {Id. at 471 '2-12:12.)
Following the adoption of SB 1256, the VDOB launched a statewide pre-election
campaign informingvoters of the photo identification requirement. This included
sending 86,000 postcards to persons on the active voter list who DMV records reflected
possessed no DMV-issued ID and would likely need a photo ID to vote under the new
law. {Id. at 1474:20-75:6, Feb. 26, 2016 (Test, of Edgardo Cortes).) This excluded
certain regular absentee voters who would not need photo ID to cast an absentee ballot.
To educate local electoral board members, general registrars, and poll workers, the
VDOE instituted trainingprograms and issued handbooks and procedural guides. {Id. at
1471:5-14; id. at 940:5-24, Feb. 25, 2016 (Test, of Myron McClees).)
Matthew J. Davis ("Davis"), the Chief Information Officer for the VDOE, also
testified that his agency employed billboard ads, radio, and Facebook, to acquaint voters
with the recently enacted identification requirements. {Id. at 1006:15-07:24.) The
VDOE contracted with a marketing agency to assist in developing an outreach strategy.
With the agency's assistance, the VDOE distributed over 500,000 fliers and posters to
registrars' offices. (Pis.' Trial Ex. 155.) There are 133 local registrar's offices in
21
Virginia.
In addition, Davis indicated that VDOE records reflected that 773 provisional
ballots were cast by voters without valid identification in 2014, as reported by 129
Virginiajurisdictions. The following year in 2015,408 provisional ballots were cast by
voters with no acceptable form of identification. In 2015, however, twenty-seven
jurisdictions failed to report the number of provisional ballots issued to voters without
identification. Statistically, this translates to .04 percent of the total ballots cast in 2014
and .03 percent in 2015. (Defs.' Trial Ex. 301.)
IV.
Plaintiffs' Evidence
To provide a historical overview of racial discrimination in Virginia, particularly
as it relates to voting rights, the Plaintiffs began their presentation of evidence with the
testimony of Dr. John Douglas Smith ("Dr. Smith"). Dr. Smith, who holds a Ph.D. in
American history from the University of Virginia and currently serves as the Director of
Humanities at Colbum Music Conservatory, was offered as an expert in Virginia history,
with an emphasis on racial discrimination. He is the author of a book entitled Managing
White Supremacy. Although he provided his impressions of the effects of contemporary
voting policies and practices, the bulk of his writings and research appears to focus on the
pre-Voting Rights Act era. Dr. Smith recounted in some detail overt measures adopted in
Virginia to inhibit minority voting from the post-Civil Warera through Massive
Resistance. In his view, following the adoption of the Voting Rights Act in 1965, the
strategy employed to suppress minority voting took a more subtle form.
According to Dr. Smith, during the post-Civil War era, African Americans initially
22
enjoyed some success in electing African Americans to the General Assembly, as well as
one to the United States House of Representatives. Followingthe adoption of the revised
Virginia Constitution in 1902, African Americans experienced a decline in political
power and influence. In Dr. Smith's opinion, with the imposition of literacy tests and
enactment of a poll tax, Virginia's African Americans were essentially disenfranchised—
and remained second class citizens until the mid-twentieth century.
Turning to the Civil Rights Era, following the decision of the Supreme Court in
Brown v. Board ofEducation, 347 U.S. 483 (1954), firmly resolved Virginia political
leaders led the Massive Resistance movement to keep public schools segregated. One
unyielding county took the extraordinary step of closing its public schools for five years
to avoid integration. Laws passed in Virginia to hinder desegregation were repeatedly
struck down by federal courts, along with Virginia's poll tax for state elections. In the
interim, the Twenty-Fourth Amendment was adopted to abolish the poll tax for federal
elections.
In discussing more contemporary times. Dr. Smithhighlighted the election of L.
Douglas Wilderas the first popularly elected African-American governor in American
history. Butas examples of continuing racial overtones in modem Virginia politics, he
pointed outthat Governor Wilder was elected by a smaller than expected margin of
victory in 1989. He also noted the Declaration of April as Confederate History and
Heritage Month by Republican governors and Senator George Allen's use of the term
"macaca" when referring to a man of South Asian descent as further evidence that race
23
continues toplay a role in Virginia's politics."* As further examples ofresidual evidence
of racial discrimination in Virginia, Dr. Smith discussed the ongoing litigation over
claims of racial gerrymandering, coupled with approximately twenty objections by the
Department of Justice to changes in Virginia's voting laws under Section 5 of the Voting
Rights Act.
In support of his conclusion that subtle remnants of discrimination remain in
Virginia politics. Dr. Smith points to the fact that Virginia has failed to elect an Afncan
American to a statewide office since Governor Wilder's election in 1989, which was
preceded by his election as Lieutenant Governor in 1985. Dr. Smith does acknowledge
that Republicans have twice nominated African-American candidates for statewide office
in recent years, while Democrats have failed to do so.^
In his final analysis. Dr. Smith concludes that Virginia's voter identification law is
consistent with the long line of actions taken over Virginia's history to suppress minority
vote. Dr. Smith appears to assume that the viewpoint of current legislators must be
infected by this inherited legacy. While Dr. Smith's testimony is informative, his broad
conclusions appear to be leavened largely by anecdotal evidence and historical inference
with scant evidentiary support. Dr. Smith neither interviewed any member of the General
Assembly nor reviewed the legislative record.
To demonstrate the burden to voters occasioned by SB 1256, Plaintiffs introduced
According to Dr. Smith, Senator George Allen received approximately sixteen percent ofthe
Afncan American vote in his 2006 failed re-election bid after he used a racial slur.
No other AfncanAmerican has pursued state-wide office in a general election as a Democrat
since 1989, except Donald McEachin. In 2001, the Democrats nominated now Senator
McEachin, an Afncan American, for Attomey General; however, he lost to Jerry Kilgore.
24
evidence from a series of individuals and local election officials who recounted
experiences prior to and during the 2014 election cycle. These witnesses uniformly
describe themselves as favoring Democratic candidates. Two of the twelve burdened
voter witnesses were African American and the other ten were Caucasian. Of the twelve,
four people actually cast votes in the 2014 or 2015 election: two timely cured their
provisional ballots, one voted absentee, and one returnedto the polls with valid
identification.
Eight other voterwitnesses made a conscious choice not to pursue othervoting
options or cure their provisional ballots, eitherbecause the winner had been declared,
they lost interest, forgot, or were angry. One ofthese witnesses, who testified that she
left the polls in frustration, indicated that no one offeredher a provisional ballot. Two
other non-voting witnesses testified that theywere unaware that they could curetheir
provisional ballots by fax or email.
In almost every case, the testifying voterwas unable to produce a valid
identification at the polls. Six of these individuals, despite being bona fide Virginia
residents, had either an out-of-state or expired driver's license.^ One had lost his voter
identification; one had no Virginia operator's license, but had a valid passport, which he
omitted to bring to the polls. One had an expired passport and anothervoter reasoned
that because the poll workers knew him, no photo ID should be necessary. Finally, two
of the voter witnesses failed to timely receive their free voter identification. Each of
^With certain exceptions for military personnel, Virginia Code §46.2-308 requires every new
resident toobtain a Virginia operator's license within sixty days ofresidency.
25
these voters was eligible to cast absentee ballots but some chose not to do so and others
were unaware that it was available.
All twelve of the allegedly burdened voters who testified expressed frustration
with their initial inability to vote without photo identification. Most testified that they
were unaware of the requirement. Each also expressed their disagreement with the need
for such unexpired identification. Some were disgruntled by the necessity to travel to the
registrar's office to cure their vote, particularly those who were disabled or elderly.
Plaintiffs also designated depositions of two affected voters as evidence. Charles
Benagh, a white male, chose not to vote in-person in 2015 even though Fairfax County
had informed him that he possessed an appropriate form of identification. (Benagh Dep.
43:1-11,48:20—50:3, Pis.' Trial Ex. 220.) Instead, he applied for and received an
absentee ballot, which he chose not to return because he did not believe that he could
mail it and have it delivered in time to be counted. (Jd,)
Mary Joanna Jones ("Jones"), an eighty-one year-old African American, attempted
to vote in-person during the 2014 general election, but she did not have an acceptable
form of identification. (Jones Dep. 11:1-14, 13:3-23, Pis.'Trial Ex. 224.) Shecasta
provisional ballot, which she cured the following Thursday. {Id. at 16:12-25, 29:5-14.)
She attempted to obtain her free photographic identification from the registrar's office
when curing her ballot, but because of a mix-up in Richmond, she did not receive her free
identification in a timely manner. {Id. at 18:4-22:20.) To accommodate her, an
individual fi-om the registrar's office came to Jones's home in September 2015 to take her
picture for the free identification. {Id. at 22:21-23:25.) She received her identification
26
before the general election and successfully cast her ballot that fall. (Jd. at 22:21-25.)
In the final analysis, none of the voter witnesses was actually denied his or her
right to vote. Admittedly, for some, the process was cumbersome. Many voters,
including a number who testified in this case, were not informed they could cast an
absentee ballot, that they could cure the provisional ballot, or obtain a free photo ID.
Others had valid identification but failed to bring it to the polls.
To further illustrate the impact of SB 1256, Plaintiffs offered the testimony of a
number of Democratic Party activists and election officials. These witnesses recounted
the difficulty in educating low income and minority voters on the requirement of photo
identification. Most of these witnesses indicated that voters in their community did not
understand the need for photograph bearing identification.
The chair and vice chair of the Henrico County Democratic Committee described
their voter outreach campaign, which was conducted in league with the SBOE. It
included phone banks, palm cards, fliers, and Facebook postings. Their efforts
specifically targeted minority and elderly voters. The vice chair described the outreach
program as successful.
The secretary of the Prince William County Electoral Board described his
community as a "battle ground" district with a fairly even minority-majority population.
(Trial Tr. 657:19-58:2, Feb. 24, 2016.) He intimated that because the local police
frequently check on members of the Latino community to ensure that they have proper
immigration identification, there may be a mistrust ofgovernment. In his view, the photo
identification requirement was unnecessary and was not well-received byvoters in his
27
county. He found the requirement particularly problematic in high turnout election years
because it contributed to longer lines.
The chair of the Alexandria Democratic Committee, who served on the Electoral
Board as well, also described her education outreach as focusing on young and elderly
voters. Particularly challenging for her was maintaining a list of colleges that have some
presence in Virginia. Since this is a prerequisite to the validity of a college identification,
she found herselffrequently having to check a schedule of approved colleges. She also
questioned the need for a photo ID to vote.
Three other Democratic operatives added their perspective. Plaintiff BarbaraLee,
from Stanton, Virginia, believed that the voter identification requirement lowered voter
turnout. She also believed that the requirement adversely impacted low income areas in
which peoplehad neitherthe time nor transportation required to obtain valid
identification. Lee, however, was only able to identify one person who, she believed,
could not vote as a result of the voter identification requirements. She also admitted on
cross-examination that she never advised this individual of her right to cast an absentee
ballot.
JeffAllen, from Alexandria, Virginia, a campaign manager, political consultant,
and Democratic field organizer, described the challenge he encountered in educating
what he described as lower turnout voters. He revealed that in explaining the
requirements ofthe voter identification law, he only mentioned the alternative ofcasting
an absentee ballot if specifically asked. He recalled encountering one bedridden voter
who, he believed, had no photo identification. Allen added on cross-examination that the
28
bedridden voter indicated no interest in casting an absentee ballot.
Plaintiff Gonzalo Aida, a member of the Richmond Democratic Committee, found
the voter ID law to be an obstacle during the 2014 elections. Aida focused his get-out-
the-vote activities on Latino communities and university campuses. He found university
students, by and large, to be unwilling to devote the necessary time to acquire photo
identification. As a poll worker, he encountered a number of people without valid photo
identification. Only a few people, however, declined to accept provisional ballots. He
did encounter some African Americans who were fhistrated and refused to accept
provisional ballots. Aida admitted on cross-examination that he was unaware of any
person who was unable to vote because of the Virginia voter identification law.
V.
Defense Witnesses
The defense called a number of Virginia election officials, most of whose
testimony is recounted in othersections of this opinion. These witnesses include Edgardo
Cortes, Commissioner of the VDOE; J. Justin Riemer, former Deputy Secretary of the
SBOE; Matthew J. Davis, Chief Information Officer, VDOE; and Donald Palmer, former
Secretary of the SBOE. The defense also introduced the testimony of Myron McClees,
Policy Advisor, VDOE.
McClees attended a number of the committee hearings on SB 1256. He
encountered considerable partisan sparring but remembered arguments in favor of the bill
as a vehicle to reduce voter fraud. McClees characterized the decision of the SBOE to
adopt a one-year expiration date for voteridentification as a compromise. McClees's
responsibilities also included educating voters on the identification requirements
29
prescribed by SB 1256. His personal voter outreach focused on low income and minority
voters. For example, he sent letters to members of the NAACP and to members of the
clergy explaining the regulations and offering further assistance. McClees was concerned
that the provisional ballot used in 2014 did not mention the availability of free voter ID.
The present provisional ballot includes such information.
The Defendants also called Cameron Quinn ("Quinn"), former Fairfax County
general registrar, as a witness. Quinn described her extensive outreach efforts to educate
the 700,000 voters in her county on post-2013 voter identification requirements. To
enhance its implementation, she employed both mobile and satellite offices to register
voters and issue free forms of identification.
Quinn readily admitted that she encountered a number of complications in
implementing the newly-enacted photo identification requirement. She recalled in July of
2014 sending a letter to Commissioner Cortes explaining problems with the mobile
system for issuing free voter identification. It required several weeks to bring the system
back online.
Quinn testified that during the 2014 election cycle, just under 500 provisional
ballots were cast in Fairfax County. Fifty of those were cast because the voter could not
present valid identification. Of that number, approximately one half were cured within
the statutorily-allotted time frame. (Trial Tr. 1718:19-19:1, Mar. 1,2016.)
VI.
The Experts' Interpretation of the Impact of SB 1256 on Minority and Young
Voters
To add an interpretative gloss to the factual evidence, each side presented an array
30
of expert witnesses, most drawn from the academic community. Their widely differing
opinions were based on statistical models shaped from surveys, public data, and academic
studies. Several experts employed analytical constructs crafted specifically for this type
of litigation, purporting to identify burdened segments of the population likely to have no
valid identification enabling them to vote. And, each expert, in varying degrees,
acknowledged Virginia's undeniable pre-Voting Rights Act history of discriminatory
votingpolicies. The experts, however, presented divergent viewpoints on the
justification for photo-bearing voter identification, as well as the motives of theVirginia
General Assembly in enacting such legislation.
a.
Dr. Allan Lichtman
To support their contention that SB 1256 was intended to discriminate against
certain groups by placing disparate burdens on votingrights, the Plaintiffs called Dr.
Allan Lichtman ("Dr. Lichtman"), a distinguished professor of history at theAmerican
University. In formulating his opinions. Dr. Lichtman applied quantitative methodology
to draw inferences from political history. His resources included scholarly books,
articles, reports, newspapers, demographics, election returns, court opinions, and
scientific surveys.^ The professor noted that he had testified many times previously as an
expert in the field of legislative intent.
^In supporting his conclusion that SB 1256 stifled minority voter turnout, he also relied on a
study conducted on a Texas voting district and a survey of200 Virginia voters. (Trial Tr.
1334:08-35:04,1405:22-06:14, Feb. 26, 2016.) Defendants' expert Dr. Owen countered that
the Texas study involved a distinctly different population base and the sample size ofthe second
survey was too narrow to be of value. {Id. at 1956:14-21, Mar. 2, 2016.) Shealso noted that
many individuals who responded to the 200-person survey ofVirginia voters that reported
lacking proper identification actually chose not to vote for other reasons. (Jd. at 1953:06-56:13.)
31
Based upon his historical and quantitative analysis, and relying on eight of the
nine factors articulated in Gingles, supra. Dr. Lichtman concluded that the Virginia voter
ID law in controversy in this case was enacted and implemented with discriminatory
intent. He further opinedthat the law was enacted not only to achieve political advantage
but also to burden the Democratic minority base. In his view, race is a fundamental
divide politically between the Democratic and Republican parties. He perceives the
political base of the Republican Party as white voters and that of the Democrats to be
African Americans. Despite significant progress in recent years, African Americans,
according to Dr. Lichtman, have a much lower income, are less likely to have a college
degree, and more likely to be unemployed than white voters. He describes the present
economic status of African Americans as a lingering effect of historical discrimination.
Dr. Lichtman conceded that much of the basis for his conclusions consists of a
mosaic of circumstantial evidence. Normally, according to Dr. Lichtman, legislators do
not openly state their intent when it is discriminatory. He highlighted the fact that votes
cast in the General Assembly on SB 1256, both in committee and on the floor, were
either party line ornear party line. It was also noteworthy to the professor that Virginia
amended the 2012 voter ID law the following year to add the photo requirement without
what he believed to be any rational basis. The statistical risk of voter fraud cannot
logically explain the addition ofa photo requirement in 2013. Dr. Lichtman was quick to
add that a Republican National Lawyers Association study was unable to identify any
cases ofvoter impersonation fraud inVirginia. Also significant to the professor in
informing his opinion was Senator Obenshain's opposition to the SBOE's decision to
32
allow expired identification. He also suggested that members of the General Assembly
should have known about academic studies showing that voter identification laws have a
disparate impact on African Americans.
In commenting on the so-called "Senate factors" relied upon in Gingles, Dr.
Lichtman drew particular attention to several factors which he contends are applicable in
Virginia. With respect to subtle racial appeals in campaigns, Dr. Lichtman mentioned a
disparaging racial comment made by Senator George Allen during his re-election
campaign in 2006 and arguably racial cartoons attributed to Republican sources.
Although he presented no information about the number of African Americans running
for public office in Virginia, he considered the fact that L. Douglas Wilder was the only
Afncan American elected to statewide office and that Afncan Americans are
underrepresented in the General Assembly. He also cited the legislature's rejection of a
proposed constitutional amendment enabling the automatic restoration of votingrights to
nonviolent former felons. Lastly, Dr. Lichtmanwas of the opinion that the failure of the
General Assembly to expand Medicaid was an example of elected officials notbeing
responsive to the needs of Afncan Americans. He offered no explanation as to how
Virginia would absorb the cost.
While Dr. Lichtman conceded that evidence of actual suppression is difficult to
unearth, he steadfastly disagreed that other states which passed a strict voter ID law did
so withouta latent motivation to suppress minority vote. To bolster his conclusion, he
elaborated by sayingthat of the fourteen states which passedvoter identification laws
after 2008, most had Republican control of the legislature. In two of those states,
33
according to Dr. Lichtman, the legislatures overrode the veto of a Democratic governor.
Rhode Island was the one state that enacted voter ID with a Democratic state legislature
and an Independent governor.
b.
Testimony of Dr. Lorraine Minnite
To provide an assessment of the frequency of voter fraud in Virginia, the Plaintiffs
called Dr. Lorraine Minnite ("Dr. Minnite"), an associate professor at Rutgers University,
Department of Public Policy and Administration. Dr. Minnite was received as an expert
in the field of American election law and voter fraud. She is the author of a book entitled
The Myth of VoterFraud. After surveying all available information and statistics. Dr.
Minnite concluded that voter fraud is rare in Virginia. Her investigation found no
reported cases of voter impersonation fraud in Virginia in recent elections. For the
purpose of her analysis, she adopted the definition of voter fraud as the intentional
corruption of the voting process by voters. Under her interpretation, the deception by the
voter has to be intentional.
Dr. Minnite's findings are based on a combination of national and Virginia state
data. She obtained prosecution statistics from the United States Department of Justice,
sent surveys to 2,700 district attorneys in the United States, and wrote letters to every
state attorney general and secretary of state. She also requested similar information from
all of Virginia's Commonwealth's Attorneys. However, she received responses from
only nineteen of over 100 Virginia prosecutors. Some of the Commonwealth's Attorneys
responded that they had received complaints of fraud-related activities by voters, but
none appeared to result in a formal prosecution. Similarly, she received information from
34
the Virginia State Police reporting convictions for election law violations. On review,
she concluded that most of these convictions were for illegal voting activities, but not
fraudulent voter impersonation.
In harvesting information concerning the incidences of voter fraud in Virginia, she
read 647 relevant news articles. She gleaned from these articles that there were
approximately sixty cases of illegal voting by felons, but none involved actual voter
impersonation.
Dr. Minnite confined her statistical analysis to actual convictions and not reports
of alleged violations to law enforcement. (Trial Tr. 770:16-71:08, Feb. 24,2016.) She
also assumed in her study that if there was credible evidence of voter fraud, the
prosecutor would have formally brought charges. She had no way of determining what
criteria prosecutors may have used to determine whether it was appropriate to pursue an
indictment or merely resolve the matter informally. Her statistics also did not capture
inadvertent voting in the wrong precinct.
Although Dr. Minnite voiced the opinion that there was insufficient evidence to
conclude that voter fraud is a rational justification for photo ID laws, she admitted that
such laws could prevent voter impersonation fraud. {Id. at 796:22-97:2.)
Dr. Minnite
also hastened to add that while she had testified previously in other cases, and it was
difficult to determine whether voter ID requirements suppress voter turnout, sufficient
information may be available now to conduct a more accurate scientific or statistical
analysis. She explained that her previous reluctance to draw any conclusion on
suppression of voter turnout was based on the number of varying factors that influence a
35
voter's decision during any particular voting cycle. {Id. at 870:12-71:20.)
Dr. Minnite also testified that she disagreed with other experts' mterpretations of
the findings and recommendations of the Carter-Baker Commission. She characterized
the findings of the Commission as principally animated by a desire to instill voter
confidence. In fact, she admitted that the Carter-Baker Commission placed greater
weight on theperception that voter ID laws enhance public confidence in the integrity of
the electoral system than the actual number of reportedvoter fi*aud cases. Dr. Minnite
also suggested that the Commission's recommendation of requiring voter ID did not
appearto evolve fi-om any significant data base or careful study. She explainedthat the
Commission placed significant reliance on a study of voter fi-aud in Wisconsin. In her
opinion, the results of that investigation revealed that the problems which surfaced were
primarily administrative, rather than criminal.
Her ultimate conclusion was that while voter impersonation fi-aud is not non
existent, it occurs too infi-equently to constitute a rational basis for adoption of the voter
identification law enacted in Virginia.
c.
Testimony of Dr. Jonathan Redden
In an attempt to determine the number and types of registered voters who may
have an acceptable form of identification comportingwith the requirements of SB 1256,
Plaintiffs presented the testimony of Dr. JonathanRodden ("Dr. Rodden"), a wellrespected professor at Stanford University. Dr. Rodden was received as an expert in
political science, particularly the use of geospatial quantitative methods. Dr. Rodden
employed three analytical constructs to form his conclusions as to which voters would
36
likely have proper identification. The underlying methodology entailed using geospatial
mapping to plot each voter from the voter file maintained by the SBOE to determine
where each registered voter resided. He relied upon data from the DMV to determine
who would have a valid DMV issued photo identification and information from the
Commonwealth to ascertain who would have a valid free voter identification card. He
also utilized several assumptions, including his belief that anyone who lived ona military
reservation had an acceptable military photo identification card. To the extent possible,
he attempted to identify individuals from the voter file that may be students based upon
age, gender, and address. Dr. Rodden then gathered data using three analytical
techniques. These included homogenous blockanalysis, ecological inference analysis,
and Catalist estimate.
The homogenous block analysis takes individuals who self-identify as a certain
race and places them in a precinct with other individuals who self-identify as being of the
same race. The ecological inference analysis, a somewhat esoteric technique developed
for voting rights litigation, operates on a census data platform. Relying on census data,
the number of individuals of a certain race in each individual block can be determined.
Dr. Rodden contends that the number of individuals in that particular block can be
identified from known data. What is unknown, however, is how the race statistics match
with the identification data. The ecological inference analysis attempts to take that
information base and puts statistical bounds on each block to determine how much of
each group most likely has appropriate identification. Dr. Rodden maintains that this is
done by ruling out impossible combinations. The Catalist estimate uses an individual's
37
full name, birth date, and associated geographical data to make an estimate of that
person's race. Dr. Rodden contends that its estimates are close to ninety percent
accurate, with one exception. The estimate is more likely to misclassify an African
American as white than vice versa.
Based upon the results of Dr. Rodden's geospatial quantitative analysis, he
concluded that identification possession rates of registered voters are higher among
whites than African Americans and Hispanics, higher among middle aged and older
voters than young voters, and higher in Republican precincts than Democratic precincts.
Dr. Rodden limited his quantitative analysis to the years 2012, 2014, and 2015. His
analysis revealed that over time there was an increase in identification possession rates.
He attributes this, however, to a decline in the number of registered voters, rather than an
increase in actual possession of identification among the voters.
Based on his homogenous block analysis. Dr. Rodden estimated that in 2012,
between seventeen and eighteen percent of African Americans lacked DMV
identification, compared to nine and eleven percent ofwhites. ^ This statistic reflects
only those individuals who lack a form of identification issued by the DMV or the free
voter ID offered at local registrars' offices. When Dr. Rodden used a more inclusive
measure of identification, which encompassed individuals residing on military bases who
probably have a form of military identification or those meeting his criteria for a probable
student, the rates of non-possession dropped significantly. Under the inclusive
Q
Dr. Thornton critiqued Dr. Rodden's approach by noting that the more heterogeneous the
block's population, the higher the potential for inaccuracy. (Trial Tr. 1755:23-56:9, 1758:1824.)
38
identification analysis, the number of whites without proper identification dropped to
between 3.2 and 3.9 percent, for African Americans between 5.4 and 6.1 percent, and
between 5.1 and 6.3 percent for Hispanics for the year 2015.
The ecological inference analysis yields results essentially parallel to those from
homogenous blockanalysis. He estimates that in 2012, approximately 17.5 percent of
Afncan Americans and 9.5 percent of whites lacked either DMV identification or free
registrar-issued identification. Using the more inclusive analysis. Dr. Rodden estimates
that approximately 3.4 percent of whites, 5.7 percent of African Americans, and 6.7
percent of Hispanics lacked appropriate identification in 2015.
Turning to the Catalist estimate. Dr. Rodden estimates that approximately 4.1
percent of whites, 5.4 percent of African Americans, and 5.6percent of Hispanics lacked
some form of valid identification in 2015. Dr. Rodden also testified that in his opinion,
younger voters were less likely to have valid forms of identification than older voters.
Within the age group twenty-five to thirty, approximately eighty-five percent of
individuals have an appropriate form of identification. This same rate applies to
individuals intheir thirties. This rate gradually increases and eventually plateaus for
individuals above the age of fifty. Virginians over the age ofseventy-five typically have
a declining rate of possession of acceptable identification.
Using precinct data acquired from the Commonwealth of Virginia, Dr. Rodden
opined that there is a correlation between precincts that gravitate toward Democratic
candidates and residents without valid voter identification.
Dr. Rodden admitted on cross-examination that the absence ofhard data required
39
him to make assumptions with respect to the number of students and persons with
military identification. He also acknowledged thathis analysis did not include
unquantifiable rates ofpossession ofsuch other forms ofvalid identification as passports,
tribal, and government or employment-issued identification. His numbers also fail to
reflect active or retired members of the military who reside offbase and possess
appropriate identification. (Trial Tr. 584:1-85:8, Feb. 24, 2016.)^
d.
Testimony of Dr. Janet Thornton
To counter the opinions of Plaintiffs' experts, the Defendants offered the
assessment of three well-credentialed political scientists. Each identified what they
believedto be flaws in the data, logic, and conclusions offered by Plaintiffs' academic
experts.
Dr. Janet Thornton ("Dr. Thornton"), an expert in the field of economics and
applied statistical analyses, is the managing director of the Berkley Research Group. She
has been an economist and applied statistician for over thirty years. She was engaged in
this case to review the expert opinions of Dr. Rodden and Dr. Lichtman, with particular
emphasis on the allegation that minorities, specifically Afi-ican Americans and Hispanic
voters, as well as young and Democratic voters, would be adversely affected by SB 1256.
With respect to Dr. Rodden's geospatial analysis. Dr. Thornton was able to
identity a variety of acceptable forms of identification not included in Dr. Rodden's data
base. This is particularly true, in her view, with respect to military personnel and
^Defendants highlight in their post-trial briefthat "almost one-halfofplaintiffs' fact witnesses
who testified they infact possessed aform ofphoto ID acceptablefor voting in Virginia .,.
would have been classified as 'No ID' by Rodden." (Defs.' Post-Tr. Br. 26-27, ECFNo. 212.)
40
students. Initially, she points out that Dr. Rodden failed to accurately identify the number
ofmilitary personnel in Virginia. He identified approximately 32,500, while the number,
according to her research, is closer to 129,000. (Defs.' Trial Ex. 301.) Dr. Rodden's data
failed to include military personnel ortheir spouses living offbase. (Trial Tr. 585:17-20,
Feb. 24, 2016 (Test, ofJonathan Rodden).) Moreover, there are approximately 830,000
veterans in Virginia, many of whom have identification enabling Veterans
Administration services. In Dr. Thornton's opinion, the failure of Dr. Rodden to
accurately include the number of military personnel inVirginia undermines the integrity
ofhis analysis. In her assessment. Dr. Rodden may have failed to include as many as
seventy-five percent or more of Virginia-based military personnel. {Id. at 1730:8-16,
Mar. 1, 2016.)
Dr. Thornton took issue with Dr. Rodden's estimate of the number of students in
Virginia. Dr. Rodden's projected number was 467,000. Dr. Thornton's research
revealed roughly 600,000 students. (M at 1734:24-35:2, 1736:2-37:22.) She also
faulted Dr. Rodden for assuming that individuals who had passports or identification
issued by the Veterans Administration, federal government, state and local governments,
nursing homes, or high schools, would most likely also have a driver's license. Dr.
Thornton testified that this was an unrealistic assumption that skewed the data that Dr.
Rodden relied upon. Dr. Thornton stressed that it is important to keep in mind that
Virginia has a higher proportion of government workers and military personnel than does
the United States generally, a fact lost in Dr. Rodden's analysis.
In construing the data mined by Dr. Rodden, Dr. Thornton agreed with a number
41
of conclusions Dr. Rodden derived from his specific information base. She conceded that
African Americans are slightly more likely than whites to lackappropriate identification.
Dr. Rodden concluded, in Table 2 of his Report, that 94.6 percent of African Americans
and 96.8 percent of Caucasians have identification meeting SB 1256 standards. (Pis.'
Trial Ex. 209.) Dr. Thornton also concurred that Dr. Rodden's ecological inference
demonstrated that roughly 3.41 percent of Caucasians lack appropriate identification, as
compared to 5.59 percent of African Americans. The results of the ecological inference
study revealed that roughly 93.3 percent of Hispanics have appropriate identification. Dr.
Thornton did not take issue with the statistics based upon Dr. Rodden's data. While Dr.
Thornton concurred in Dr. Rodden's conclusions based upon his statistics and his
assumptions, she continued to question their reliability because he failed to account for
persons with other forms of identification.
e.
Testimony of Dr. Karen Owen
The Defendants' next expert. Dr. Karen Owen ("Dr. Owen"), focused her
testimony on the conclusions offered by Dr. Lichtman. Dr. Owen is an assistant
professor at Reinhardt University who was received as an expert in political science,
specializing in southern American politics and government. Her opinion of Dr.
Lichtman's report, methodology, and conclusions was predicated not only on her
experience and expertise in the field of Southern politics, but also scholarly works, public
documents, and records and information available on the SBOE's and General
Assembly's websites.
Dr. Owen testified that public opinion polling, specifically those conducted by
42
Pew and Quinnipiac, reflected overwhelming support for photographic identification by
voters. (Trial Tr. 1924:7-25:17, Mar. 2, 2016.) Based on her review of the Carter-Baker
report, she concluded that the Virginia ID law permitted more forms of identification than
recommended in that report.^® In her opinion, public perception is a significant factor in
legislative action. Even if voter impersonation fraud is rare, it is not unreasonable for
legislatures to conclude that it is sound public policy to adopt legislation assuring the
public of the integrity of theirelectoral system. Furthermore, adopting laws to deter
corruption of the electoral system before it occurs is simply an exercise of wise judgment.
{Id. at 1932:7-12.)
While Dr. Owen conceded that she found no evidence of widespread voter
impersonation fraud nationally, and no concrete examples from Virginia, she disagreed
with Dr. Lichtman's conclusion thatthe Virginia General Assembly enacted the voter ID
law with the intent to discriminate against minorities. From herreview of therecord, she
found no evidence of an intent to discriminate. In fact, her study revealed a number of
alternative, legitimate reasons why the Virginia General Assembly passed the 2013 voter
identification law. This included public opinion favoring such legislation, a public
perception ofpotential voter fraud, promoting confidence in the integrity of the electoral
system, and sound public policy in preventing future acts of voter fraud. {Id. at
1920:20-21:17.)
Dr. Owen was asked on cross-examination about an editorial penned bythe authors ofthe
Carter-Baker report. President Carter and Secretary Baker concluded that the voter identification
law enacted in Georgia was discriminatory because itdid not originally include aprovision for a
free photo identification. The 2013 Virginia legislation did infact provide for a free form of
photo identification for voters.
43
f.
Testimony of Dr. Daniel J. Palazzolo
The Defendants' final expert witness was Dr. Daniel J. Palazzolo, a University of
Richmond professor with expertise in Virginia legislative and congressional politics and
history. With respect to Dr. Smith's characterization of pre-1965 Virginia political
history. Dr. Palazzolo essentially concurred that the period was tainted bydiscriminatory
legislation andpractices. Obstacles, such as a poll tax and spurious literary tests, were
tactics calculated to suppress African American voting rights. These ploys were fostered
by a white-dominated, single political party system which enjoyed substantial public
support.
Following the adoption of the Voting Rights Act in 1965, with enforceable
safeguards against racial barriers, Dr. Palazzolo perceives a significant course correction.
Incontrast to Dr. Smith, Dr. Palazzolo views contemporary history differently. As for
post-1965 voter participation among African Americans, Dr. Palazzolo discerns a clear
trajectory toward greater inclusion. In contrast to the so-called Jim Crow era, in recent
years, Virginia has had a vibrant, competitive, two-party system supervised by a nonpartisan electoral administration.
Dr. Palazzolo also took issue with Dr. Smith's conclusion that opposition to the
Voting Rights Act by the Virginia Congressional Delegation in 1965 supports Dr.
Smith's opinion that conscious voter suppression is a continuing legacy inthe Old
Dominion. Dr. Palazzolo highlights the Virginia Congressional Delegation's unanimous
support for the 2006 reauthorization ofthe Voting Rights Act as compelling evidence of
44
an evolving acceptance of a more racially inclusive voter base.'^
As further evidence of a more diverse and vibrant political base, Dr. Palazzolo
noted that Virginia was the first state since the enactment of the Voting Rights Act to
elect a black governor. In fact, L. Douglas Wilderhad also previously served as the
state's lieutenant governor. Furthermore, as Dr. Smith alsopointed out. Republicans
have nominated two African American candidates for statewide office, while Democrats
10
have nominated none.
Unlike Dr. Smith, part of Dr. Palazzolo's portfolio of academic responsibilities is
following the Virginia General Assembly. In fact, he directs the University's internship
program in the General Assembly. In Dr. Palazzolo's opinion, Dr. Smith's conclusion
that the legislature's intent in enacting the 2013 voter ID law was suppression of minority
vote was based on pure speculation unmoored to any factual basis. He testified that it
was equally plausible that the General Assembly was simply responding to the public's
perception that Virginia needed a stronger voter identification law. He also noted that
Lieutenant Governor Bill Boiling, a Republican, supported a Democratic amendment to
delay implementation of the law until after the 2013 election to allow voters to be
informed of the additional identification requirement. Other acts of the General
Assembly further demonstrated an absence of intent to suppress minority voters. This
included allowing for a broad array of acceptable forms of identification andthe issuance
of a free voter identification card by the Commonwealth.
One member of the Virginia Congressional Delegation did not vote.
Dr. Palazzolo omitted to mention that Donald McEachin was nominated by the Democrats for
Attorney General in 2001, However, as previously mentioned, he was not elected.
45
In his final conclusion, Dr. Palazzolo admitted that reasonable partisan minds
could differ as to whether or not there was an actual need for photograph bearing voter
identification. He also indicated that the legislative history and public record are
insufficient to draw a defensible opinion that any member of the General Assembly voted
for SB 1256 with an intent to suppress minority vote.
VIL
a.
Weighing the Evidence in the Context of Plaintiffs' Claims
Count I: Alleged Violation of Section 2 of the Voting Rights Act,
codified at 52 U.S.C. § 10301(a)
While this lawsuit wages a comprehensive, multi-faceted constitutional challenge
to the Virginia voter ID law, the focal point of this litigation is Plaintiffs' claim under
Section 2 of the Voting Rights Act. From an evidentiary perspective, this vote-denial
claim is closely allied with Count II, alleging violations of the First Amendment and the
Equal Protection Clause of the Fourteenth Amendment.
As earlier explained, the determinative inquiry under Section2 is whether SB
1256 inhibits equal participation byprotected classes in the electoral process. More
sharply focused. Plaintiffs must demonstrate that the voting requirement at issue—^the
presentation of photo ID—diminishes the opportunity of protected classes from
participation in the political process. See League of Women Voters ofN.C., 769 F.3d at
238 (quoting 52 U.S.C. § 10301(b)).
The legal analysis under Section 2 ofthe Voting Rights Act does not hinge on
proof on intentional discrimination; discriminatory results are sufficient. Chisom, 501
U.S. at 404. The import of the statute is well-captured in Gingles: "The essence of a
46
[Section] 2 claim is that a certain electoral law, practice, or structure interacts with social
and historical conditions to cause an inequality in the opportunities enjoyed by black and
white voters to elect their preferred representatives." 478 U.S. at 47. The teachings of
Gingles yield two critical elements of proof First, that the challenged statute—^here SB
1256—imposes "a discriminatory burden on members of a protected class, meaning that
members of the protected class have less opportunity than other members of the
electorate to participate in the political process." League of Women Voters ofN.C., 769
F.3d at 240 (internal quotation marks and citation omitted). Secondly, there must be
proof of some causal link to "'social and historical conditions' that have or currently
produce discrimination against members of theprotected class." Id. (quoting Gingles,
478 U.S. at47).*^ Section 2 ofthe Voting Rights Act clearly proscribes racial or ethnic
discrimination "but does not require states to overcome societal effects of private
discrimination that affect the income or wealth of private voters." Frank v. Walker, 768
F.3d 744, 753 (7th Cir. 2014), cert denied 135 S. Ct. 1551 (2015).
The Supreme Court in Gingles suggested a number of potentially relevant factors
to be considered in evaluating Section 2 claims. Section 2 vote denial claims should not
be viewed in isolation, but in light of the totality of circumstances. See Gingles, 478 U.S.
at 45.
There is no serious dispute in this case that the Commonwealth of Virginia, like
many states, has a regrettable history of discriminatory policies and practices designed to
Some consideration of pre-1965 social and historical conditions is an appropriate starting
point, but this Courtwill accord greaterweight to more contemporary patterns and practices. See
McCleskey v. Kemp, 481 U.S. 279, 298 n.20 (1987).
47
suppress voting within the black community. The Fourth Circuit inLeague of Women
Voters ofNorth Carolina identified polarized voting as a potentially important analytical
factor. 769 F.3d at 240. The evidence confirmed the commonly held assumption that
African American voters tend to gravitate toward the Democratic Party. In recent years,
however, an increasing number of African Americans have run for statewide office on the
Republican ticket. As voters embrace individualized views on a wider diversity ofissues,
political lines are increasingly blurred.
The evidence, however, clearly revealed a progressive pattern ofpost-Voting
Rights Act remediation. Forexample, the Court heard testimony indicating that voters
may now apply for an absentee ballot online and that the state allows electronic
transmission of voter registration materials from the DMV to local registrars to
streamline registration of eligible voters. Virginia also provides voters with a
photographic ID free of charge. In implementing SB 1256, the SBOE adopted a
regulation defining a validID as one that has not been expired for more than a year. In
doing so, the SBOE expanded on its previous guidance by lengthening the applicable
time frame from one month to twelve months. In recent years, Virginia has taken
aggressive steps to eliminate barriers and ensure that all citizens have an equal
opportunity to vote.
While SB 1256 may have added a layer of inconvenience to the voting process, it
appears to affect all voters equally. None of the voterwitnesses produced by Plaintiffs
identified any legal obstacle inhibitingtheir opportunity to vote. Persons without valid
photo identification were able to cast provisional ballots and cure them bypresenting
48
proper evidence within three days, oralternatively, if they were disabled, submitting an
absentee ballot. Of the fourteen burdened voter witnesses who testified, either in court or
by deposition, three were African American and the other eleven were Caucasian. Five
of these fourteen individuals were actually able to cast votes in either the 2014 or 2015
election, and the remainder made a conscious choice not to pursue other voting options
when they were initially unable to produce valid identification. Three of the non-voting
witnesses were unaware that they could cure theirprovisional ballot by fax or email, or
castan absentee ballot. Unquestionably, the transition to photograph bearing
identification experienced a fair number of glitches. This was primarily attributed to
inadequate training of registrars and poll workers on the nuances of casting a provisional
ballot under the new law. Some voters received inadequate or incorrect instructions from
uninformed election officials.
However, none of the voter witnesses who testified in
this case were actually denied their right to vote.
As the United States Court of Appeals for the Seventh Circuit explained in Frank
V. Walker, "the inconvenience of making a trip to the [department of motor vehicles],
gathering the required documents, and posing for a photograph surely does not qualify as
a substantial burden on the right to vote." 768 F.3d at 748 (alteration in original)
(quoting Crawford, 553 U.S. at 198).
As a foundation for their claim of lingering racial discrimination in the electoral
process, Plaintiffs called a number of academic experts who expressed their view that the
While regrettable, any isolated mistake by an election official does not render the entire statute
unconstitutional.
49
Virginia photo ID requirement was intended to suppress minority vote. Aside from
dusting off bits of superannuation history from the Jim Crow era, such as literacy tests
and poll taxes, the evidence offered to support these conclusions were the number of
African Americans running for public office; the disproportionality of minority members
of the Virginia General Assembly; an isolated derogatory comment by a candidate for the
United States Senate; the practice of a county police department to inquire about the
immigration status of Latinos; Donald Trump's narrow victory in the 2016 Republican
primary; unidentified, unflattering cartoons of the President of the United States and
criticisms of his policy positions; the failure of the Virginia General Assembly to expand
Medicaid funding, notwithstanding its budgetary impact; and ongoing litigation over
racial gerrymandering.
Most of Plaintiffs' experts deduced that discrimination was the motive for SB
1256 because, in their opinions, there was simply no rational basis to require the
presentation of photo-bearing identification to vote. In their view, the General
Assembly's proffered rationale was tenuous and simply a veiled vehicle to inhibit African
Americans, Latinos, and young voters from participating in the electoral process.
Embracing this explanation for the more stringent identification requirements of SB
1256, Dr. Allan Lichtman testified that legislatures rarely expose their racial or ethnic
bias inthe legislative record. Such intent, in his opinion, can only be proven
circumstantially. (Trial Tr. 1123:24—24:9, Feb. 25, 2016.) In addition to the evidence
described above, Plaintiffs highlight the allegedly unprecedented introduction of SB 1256
one year after the General Assembly adopted a non-photo identification bill in 2012
50
without any intervening circumstances warranting such action. Additionally, SB 1256
necessitated an elaborate and costly voter re-education program.
Plaintiffs also highlight the action of a single Republican senator who not only
introduced SB 1256, but also attempted to persuade the SBOE to require that valid
identification include only unexpired documents. The SBOE, whose then existing policy
prohibited identification expired more than thirty days, compromised by adopting a
regulation defining valid as not expired over twelve months. Even assuming, arguendo,
that a single Republican senator had a latent motive to effect minority vote,^^ such motive
could not on the record at hand be imputed to the other Republican senators, along with
one Democrat and one Independent in the House of Delegates, who voted for the bill.
How many affirmative voters would be necessary to prove that a legislative body adopted
a measure with discriminatory objective? That question remains unanswered!
Approaching the intent of the legislature from a different angle. Plaintiffs contend
that fraud prevention was simply a transparent ruse. Plaintiffs' expert. Dr. Minnite,
conducted a comprehensive study of impersonation-type voter fraud convictions and
concluded that such criminal activity is rare. The balance of the evidence seemed to
support this conclusion but equally revealed a public perception that itwas a legitimate
concern.
Even though evidence of actual voter impersonation-type fraud was scant, there
was considerable public support for preemptive legislation deterring such criminal
activity. The Supreme Court in Crawford acknowledged that photo identification is an
The Court does not imply that there was evidence that Senator Obenshain had such motives.
51
effective method of enhancing the integrity and reliability of the electoral process. 553
U.S. at 204. Although several members of the General Assembly testified that they
suspected that Republicans who unanimously supported SB 1256 were seeking tactical
advantage rather than voter integrity, there was no evidence to elevatethis impression
beyond suspicion. In Crawford, the Supreme Court noted in the equal protection context,
"if a nondiscriminatory law is supported by valid neutral justifications, those
justifications should not be disregarded simply because partisan interests may have
provided one motivation for the votes of individual legislators." Id. at 204. The Court
further found that "[t]here is no question about the legitimacy or importance of the State's
interest in counting only the votes of eligible voters
While the most effective method
ofpreventing election fraud may well be debatable, the propriety of doing so is perfectly
clear." Id. at 196.
To further demonstrate the potentially suppressive effect of SB 1256 and to bolster
their claim of racially discriminatory impact, Plaintiffs offered extensive demographic
and geospatial quantitative analysis. Relying on these statistical models. Dr. Rodden
provided an educated estimate of the number of potential voters in Virginia without a
valid identification issued by the DMV, local registrars' offices, Virginia-based colleges,
or United States military. Across his three analytical constructs using his inclusive
analysis. Dr. Rodden concluded for 2014 that approximately 9.5 percent of Afi-ican
Americans and 10 percent of Latinos lacked a valid form of identification. For 2015, he
estimated that each figure decreased by nearly halfto roughly 5.5 percent of African
Americans and roughly 6 percent of Latinos.
52
While Dr. Rodden undoubtedly based his calculations on the best available data
and employed widely accepted methodology, he conceded that his statistics for military
personnel and students were based on what he believedto be reasonable assumptions.
Dr. Rodden's analysis relied on seemingly trustworthy demographic data from which he
identified potential minority voters without certain forms of valid identification.
Although the integrity of Dr. Rodden's statistical calculations was unchallenged, its
weight in demonstrating denial of equal voter opportunity is questionable.
Dr. Thornton, a defense expert, pointed out that Dr. Rodden's analysis did not
include individuals with passports, tribal or government-issued identification, including
elderly military veterans who may not possess a valid Virginia driver's license. (Trial Tr.
1743:8-44:13, Mar. 1, 2016.) Dr. Thornton also noted that Dr. Rodden appeared to
exclude military personnel living off base, which he admitted was a difficult statistic to
accurately capture. Dr. Thornton also testified that in her view. Dr. Rodden had
underestimated the number of military personnel and students residing in Virginia. (Id. at
1727:3-28:21, 1733:19-23.)
Positing that Dr. Rodden's analysis is sound, his findings would still not provide a
reliable picture of demographic groups allegedly denied an equal opportunity to vote.
Aside from his inability to include certain forms of valid identification, on close
examination it is also apparent that his statistical calculation makes no adjustment for
persons prohibited by law from voting, such as some convicted felons or non-U.S.
citizens. Even more importantly, his analysis also excludes people who simply have no
53
interest in voting.
Dr. Thornton acknowledged that Dr. Rodden's statistical analysis supports the
conclusion thatAfrican Americans, as a demographic block, are by a slim statistical
margin less likely to have a form of valid identification. Neither this statistical
conclusion nor Dr. Rodden's analysis supports a reliable factual finding that African
Americans or Latinos are denied an equal opportunity to participate in the electoral
process. Nothing presented supports the conclusion that minorities are not afforded an
equal opportunity to obtain a free voter ID. As described by numerous witnesses during
the course of the trial, eligible voters do not need to present any independent
documentation to obtain a free voter form of identification under Virginia Code § 24.2643 and its implementing regulations. The statute simply requires that a registrant
provide her name, address, birthdate, and social security number and sign the registration
form swearing that the information provided is true and correct.
In assessing the viability of Count I, alleging a violation of Section 2 ofthe Voting
Rights Act, this Court again draws on the wisdom ofthe Supreme Court in reviewing a
similar photo identification law challenged in Crawford. Justice Stevens, speaking for
the Court, perceptively described the task athand in the present case. Significantly, he
noted that the statistical evidence in the record did not provide the court with the number
In assessing the actual impact ofthe photo ID requirement, it isimportant tonote that only
41.6 percent of registered voters voted in 2014, and 29.1 percent in 2015. (Defs.' Ex, 324
(Summary ofVirginia Registration &Turnout Statistics chart from the VDOE website).) To the
extent that Plaintiffs objected to Defendants' Exhibit 324, the Court "may take judicial notice of
official government reports and statistics." United States v. Cecil, 836 F.2d 1431,1452 (4th Cir
1988) (citing FRE 902).
54
of registered voters without photo identification. Crawford, 553 U.S. at 200. Justice
Stevens then concluded
Petitioners askthis Court, in effect, to perform a unique balancing analysis
that looks specifically at a small number of voters who may experience a
special burden under the statute and weighs their burdens against the
State's broad interests in protecting election integrity. Petitioners urge us to
ask whether the State's interests justify the burden imposed on voters who
cannot afford or obtain a birth certificate and who must make a second trip
to the circuit court clerk's office after voting. But on the basis of the
evidence in the record it is not possible to quantify either the magnitude of
the burden on this narrow class of voters or the portion of the burden
imposed on them that is fully justified.
Id.
This Court finds itself in a similar posture. Finding the evidence insufficient to
support Plaintiffs' claim that SB 1256 has denied Afncan American, Latino, and young
voters an equal opportunity to participate in the political process and to elect
representatives of their choice. Count I will be dismissed. Plaintiffs have not shown by a
preponderance of the evidence that the statute has an adverse disparate impact on Afncan
American and Latino voters, imposes a discriminatory burden on those protected classes,
or causes anyone to have less opportunity than others to participate in the political
process.
b.
Count II: Undue Burdens on the Right to Vote and Disparate
Treatment of Voters Without a Rational Basis
The essence of Count II is well-captured in paragraph 110 of the Amended
Complaint,
The voter ID law imposes burdens on voters generally and severe burdens
on African-American, Latino, young, poor, and Democratic voters, as well
as the class of voters who lack an ID that can be used for voting. Given
55
that the law does not materially benefit Virginia or plausibly further any
other permissible interest, the burdens imposed by the voter ID law
outweigh the benefits of the law and it must therefore be invalidated under
the Equal Protection Clause.
(Am. Comply 110.)
All parties in this case agree that this Court's review of SB 1256's
constitutionality under the First Amendment and Equal Protection Clause of the
Fourteenth Amendment is guided by the Anderson-Burdick balancing framework. As
noted above, the Supreme Court has explained that in applying the framework, a court
evaluating a constitutional challenge to an election regulation must "weigh the asserted
injury to theright to vote against the precise interests put forward by the State as
justifications for the burden imposed by its rule." Crawford, 533 U.S. at 190 (quoting
Burdick, 504 U.S. at 434) (internal quotation marks omitted).
To require that every voting regulation be subjected to strict scrutiny, "andto
require that the regulation be narrowly tailored to advance a compelling state interest,...
would tiethe hands of States seeking to assure that elections are operated equitably and
efficiently." Burdick, 504 U.S. at433. Accordingly, "when a state election law provision
imposes only 'reasonable, nondiscriminatory restrictions' upon the First and Fourteenth
Amendment rights ofvoters, 'the State's important regulatory interests are generally
sufficient tojustify' the restrictions." Id. at434 (quoting Anderson, 460 U.S. at 788).
17
Although this sounds ofrational basis review, Crawford did not articulate a bright-line
standard for analyzinga voter ID law underAnderson and Burdick. Mindful that the Anderson-
Burdick balancing framework amounts to a sliding scale depending on the severity ofany
restriction and believing that Crawfordcontrols analysis, the Courtendeavors to hew to
Crawford's teachings by balancing any restrictions imposed by SB 1256's facially neutral
56
The Supreme Court has repeatedly recognized "that States retain thepower to regulate
their own elections." Id. at 433 (citing Tashjian v. Republican Party ofConn., 479 U.S.
208,217 (1986); Sugarman v. Dougall, 413 U.S. 634, 647 (1973)). The Supreme Court
also emphasized inAnderson as well as Burdicki\xdiX "[e]lection laws will invariably
impose some burden upon individual voters. Eachprovision of a code, 'whether it
governs the registration and qualifications of voters, the selection and eligibility of
candidates, or the voting process itself, inevitably affects - at least to some degree - the
individual's right to vote
Burdick, 504 U.S. at 433 (quoting Anderson, 460 U.S. at
788).
Although it is difficult to quantify with any degree of reliableprecision, the
Plaintiffs' evidence arguably would support the fact that African Americans and Latino
voters are slightly less likely to have valid identification than Caucasians. The burden
imposed upon them by SB 1256 would be the burden to travel to the DMV or the local
registrar's office to obtain an acceptable form of identification. Even if the record
evidence was sufficient, the number of such burdened individuals would not be sufficient
for this Courtto conclude that SB 1256 imposed excessively burdensome requirements
on any class of voters. Crawford, 553 U.S. at 202 (quoting Storer v. Brown, 415 U.S.
724, 737 (1974)). Further, it has already been established and notedthat having to take a
trip to an office and pose for a photograph does not constitute a substantial burden on the
right to vote.
The record evidence fails to support Plaintiffs' contention thatthe Virginia photo
requirements against the state's important regulatory interests.
57
ID law is arbitrary, irrational, or invidiously discriminatory, in either its enactment or
implementation. While reasonable people can debate aspects of this law, no feature of
the law is without some rational basis, including the definition of a valid ID, the ability to
obtain a fi-ee voter ID without providing underlying documentation, and the fact that IDs
issued by other states or out-of-state colleges are not accepted. Again, while debatable,
lines must be drawn, and this Court is not convinced that these elements of the regulatory
regime lack reasoned justification. No voters have been identified who have been
deprived of their opportunity to vote. Those unable to produce valid identification for in-
person voting are nevertheless able to vote by absentee or provisional ballots. Despite
arguments to the contrary, the statute's stated intention of protecting the integrity and
reliability of the electoral process serves a substantial governmental interest. Although
statistics reveal few convictions nationally for voter impersonation fraud, the evidence
has shown wide public support for adopting such legislation. Outlawing criminal activity
before it occurs is not only a wise deterrent, but also sound public policy. As the
Supreme Court reasoned in Crawford, voter identification requirements may place a
heavier burden on a limited number of persons. However, such burden is insufficient to
outweigh the state's broad interest in protecting election integrity. Crawford, 553 U.S. at
199-200.
c.
Count III: Partisan Fencing
Count III, styled "Partisan Fencing," alleges that the General Assembly enacted
the voter ID law to suppress and fence out the vote of Democrats "because of the way
they are expected to vote." (Am. Compl. 1116.) This claim necessarily assumes that all
58
voters identifying themselves as Democrats vote similarly—in other words, a straight
party line. The term "partisan fencing" is derived from Carrington v. Rash, 380 U.S. 89
(1965), and is somewhat of an aberration. It has been rarely deployed in election law
litigation thereafter. It does not appear to create a separate cause of action but may be a
useful analytical tool in evaluating First Amendment and Equal Protection Clause cases.
The circumstances in the Carrington case that form the genesis of the term
"partisan fencing" were truly unique—an attempt by the Texas legislature to deny certain
military personnel who were not permanent residents of that state from voting in any
election in Texas. Id. at 89. In the immediate case, however, no evidence was presented
that would support a credible argument that the Virginia voter identification law was
intended, or had the effect, of disenfranchising Democratic voters, much less that the text
of the law specifically targeted Democrats as the Texas law did military members. Even
if the evidence had revealed thatpartisan advantage was a latent motive in enacting SB
1256, it would not offend the First or Fourteenth Amendment.
d.
Counts IV and V: Intentional Discrimination on the Basis of Race and
Age (Respectively)
In the final two counts of the Amended Complaint, Plaintiffs allege intentional
discrimination on the basis of race and age. Count IV alleges that SB 1256 was intended
by the Virginia General Assembly, at least in part, to discriminate on the basis of race in
the voting context violating the Fourteenth and Fifteenth Amendments. Count V
contends that the voter identification law at issue places unnecessary burdens and barriers
on the right of young voters to participate in the election process in violation of the
59
Twenty-Sixth Amendment. Unquestionably, as Plaintiffs point out intheir post-trial
memorandum, "[l]egislation enacted with the intent, at least in part, to discriminate on the
basis of race [orage] in the voting context violates the Fourteenth [,] Fifteenth [and
Twenty-Sixth] Amendments." (Pis.' Post-Tr. Br. 31 (citing Vill ofArlington Heights,
429 U.S. at 264-65).)^^ Furthermore, "[d]etermining whether invidious discriminatory
purpose was a motivating factor demands a sensitive inquiry into such circumstantial and
direct evidence of intent as may be available." {Id. at 35.) In examining circumstantial
evidence to discern the intent of legislative action, courts consider its historical
background, the sequence of legislative events, departures from normal procedures, and
its disparate impact on effected populations.
Plaintiffs' evidence of intentional discrimination is animated by undisputed
evidence of pre-1965 discriminatory barriers to minority voting, arguable inconvenience
to some unidentified minority voters, the adoption of SB 1256 one year after the
legislature adopted an identification law which allowed the use of certain non-
photographic IDs, and the actions of a single state senator who introduced the bill in
question and insisted that the form of identification be unexpired. This evidence was
burnished by a number of expert witnesses offering their opinion that based on the record
at hand, they perceived no other rationale for the photo ID law except suppression of
minority vote.
152
The Court notes that there is a scarcity of caselawinvolving Twenty-Sixth Amendment claims
based on an alleged denial or abridgement of the right to vote. Yet, even accepting Plaintiffs
theory that the Twenty-Sixth Amendment essentially acts as the Fifteenth Amendment with
young voters as the pertinent class. Plaintiffs have failed to show that SB 1256 was intended,
either in its enactment or implementation, to discriminate against young voters.
60
The evidence, however, demonstrated that irrespective of statistics, a large
segment of Virginia voters thought a photo identification requirement for voting was a
prudent safeguard measure. As one expert noted, responding to public concern by
passing a law to prevent crime before it happened amounted to a reasonable action on the
part of the General Assembly. In fact, the Supreme Court agreed in Crawford. See 553
U.S. at 197. Further, voter confidence, uniformity, and fraud prevention all stood as
legitimate reasons to enact SB 1256.
Additionally, the evidence failed to show any departure from normal legislative
procedures. Instead, although ultimately passing on a near-party-line vote, the bill was
subject to robust debate from all sides. Finally, there was a complete dearth of statements
by legislators indicating any sort of discriminatory intent.
The extensive testimonial and documentary evidence offered in this case has failed
to reveal by a preponderance of the evidence that the Virginia General Assembly, a
legislative body composed of 140 Delegates and Senators, enacted the Virginia photo
identification requirement withthe intent to suppress minority and young voters.
VIII.
Conclusion
While the merits of this voter identification law, and indeed all aspects of
Virgmia's voting regime, can be reasonably debated, it remains truethat Virginia has
created a scheme of laws to accommodate all people in their right to vote. From in-
person voting, to an absentee option, to provisional ballots withthe ability to cure, and
the provision of free voter IDs, Virginia has provided all of its citizens with an equal
opportunity to participate in the electoral process.
61
Mindful thatthe Court's mission is tojudge notthewisdom of the Virginia voter
ID law, but rather its constitutionality, this Court cannot say that Plaintiffs have met then*
burden of proofin showing by a preponderance of the evidence that the Virginia voter ID
law, either on its face or in its enactment, contravenes the Voting Rights Act, the First
Amendment, the Fourteenth Amendment, the Fifteenth Amendment, or the Twenty-Sixth
Amendment. The Court finds for Defendants on all counts.
An Appropriate Order shall issue.
The Clerk is directed to file this Memorandum Orderelectronically and notify all
counsel accordingly.
It is SO ORDERED.
/s/
Henry E. Hudson
United States District Judge
Date:
Richmond, Virginia
vireini
62
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?